Choosing Your MacStadium Data Center Location
MacStadium has data centers across the United States in Atlanta, Las Vegas, and Silicon Valley as well as in Dublin, Ireland. When you select a macOS cloud resource hosted by MacStadium – compute, storage, or anything really – you will have the option of choosing which data center you would like your resource stood up in. But which location will be the best for you, and how can you make that determination?
The answer to that question will depend on both where the consumers of your resource are located and the types of work you are using your resource for.
What is latency, and how will it impact your choice of location?
Latency is the time it takes for a given command to travel from a local machine to a remote server, and the time it takes for the server to deliver its response. This time may be entirely inconsequential, or it may have a major impact on your user experience, depending on what you are using the resource for. This is because some use cases rely on continuous requests to and responses from your macOS resource, and others will require only one or very few requests for the resource to respond to. The more frequent the requests, the greater the likelihood of accumulating wait time across those requests, which can result in a less than optimal user experience.
Generally speaking, the closer your macOS resource is to your end-users, the lower the latency will be. This is because the primary driver of latency is the distance that the user’s commands need to travel to and from the remote server. However, in the case of continuous integration (CI), there will be very few requests made to your macOS resource, so there will be no opportunity for wait time to accumulate across requests. As such, the location you select for CI will likely have no real impact on the speed with which your CI jobs are executed.
Who, or what is the consumer of your macOS resource?
When considering the impact of latency on your end user’s experience, you’ll want to think about who or what will be waiting on a response from the remote server.
If you are deploying a web application, for example, your end users are humans who will be keenly aware of the time it takes for the server to respond to their every action in the browser. The same will apply to users who are making a VNC connection to make direct use of a macOS desktop environment. In cases such as these, there will be a continuous stream of requests being made to the macOS resource, which will increase the likelihood of an accumulation of wait time across requests. Choosing a data center location nearest to the consumers of your resource will reduce the potential for cumulative wait time, and any negative impact on user experience.
What types of work is your macOS resource responsible for?
Given that latency is measured in milliseconds, it may not have any real impact on the performance of your macOS cloud resource. For example, if you are running CI jobs in a macOS compute resource, the vast majority of your wait time will be dictated by the various building and testing steps that need to be carried out, and the CI job will only require a single request being made when the job kicks off. As such, you will have far more control over the speed of your CI jobs by optimizing your build environment rather than by making the “right” choice of a data center location.
If, on the other hand, you are using a VNC connection to gain access to a remote macOS desktop environment to run graphics-intensive applications, such as video editing tools, you will be making constant requests to the remote resource. In this case, you will want to be more thoughtful about the location you select when you sign up for your macOS resource. You will want to make sure to choose the location nearest to you or your team so that you can have the nearest experience possible to using a local machine.
When choosing a data center to start a subscription, keep in mind the geographic locations of all of those attempting to access the server (that could mean you, your team, or your customers). If low latency is important, the closer the servers are physically the better your connection quality will be.